EP2826254A1 - Mise en correspondance de sentiments dans un élément de contenu multimédia - Google Patents

Mise en correspondance de sentiments dans un élément de contenu multimédia

Info

Publication number
EP2826254A1
EP2826254A1 EP13708972.8A EP13708972A EP2826254A1 EP 2826254 A1 EP2826254 A1 EP 2826254A1 EP 13708972 A EP13708972 A EP 13708972A EP 2826254 A1 EP2826254 A1 EP 2826254A1
Authority
EP
European Patent Office
Prior art keywords
content item
segment
sentiment
media content
map
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
EP13708972.8A
Other languages
German (de)
English (en)
Inventor
Jehan Wickramasuriya
Venugopal Vasudevan
Chao Xu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bison Patent Licensing LLC
Original Assignee
General Instrument Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by General Instrument Corp filed Critical General Instrument Corp
Publication of EP2826254A1 publication Critical patent/EP2826254A1/fr
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28—Databases characterised by their database models, e.g. relational or object models
    • G06F16/284—Relational databases
    • G06F16/285—Clustering or classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/84—Generation or processing of descriptive data, e.g. content descriptors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/845—Structuring of content, e.g. decomposing content into time segments
    • H04N21/8456—Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments

Definitions

  • the present invention is related generally to electronic media and, more particularly, to evaluating sentiments in electronic media.
  • a media content item e.g., a movie, television program, or audio segment
  • segments of the content item are determined to be, for example, "happy,” “exciting,” “sad,” “funny,” and the like.
  • sentiments any information about the media clip may be evaluated such as the video and audio of the clip, metadata concerning the clip (e.g., close-captioning information and a description in an electronic program guide), and even social-networking responses to the content item.
  • a "sentiment map" is created that delimits segments of the content item and contains the sentiment-state keywords associated with the segments.
  • the delimitation of segments is based on the sentiment evaluation itself. Some embodiments also allow the delimiting and evaluating to be performed in real time (that is, while the media content item is being received), which can be important when social responses are included in the evaluation.
  • some embodiments include in the sentiment map an amplitude for each assigned sentiment keyword (e.g., how funny was it?) and a confidence value for the segment delimitation and for each keyword.
  • Segments may overlap, and a segment may encompass the entire media content item. (This can be useful when the content item is very short.)
  • the evaluation in some embodiments, considers information (e.g., profile and demographics) about a user actually watching the media content item. In this situation, an attempt is made to create a sentiment map personalized to this user's preferences. For example, a personalized sentiment map may tag a segment of a content item as exciting because this particular user is known to be an enthusiast for dog shows. The same segment is not tagged as exciting in a non-personalized sentiment map if the general public does not share this enthusiasm.
  • information e.g., profile and demographics
  • the sentiment map thus created can be used in a number of applications.
  • the map is considered by an advertisement broker.
  • the broker uses the map to match his advertisement offerings with segments of the media content item and can thus place appropriate advertisements at times when they would be most favorably received.
  • the advertising campaign may be directed to a primary device (on which a user is viewing the media content item) or to a companion device associated with the same user.
  • advertisement brokers can submit bids to have their advertising placed during advantageous times.
  • a particularly sophisticated system could re-evaluate the content item after a bid was accepted, the re-evaluation based on additional information that just became available (e.g., social responses to a live broadcast). If the re-evaluation shows that the original evaluation, on which the bid was based, was not very accurate, given the additional information, a refund of a portion of the bid could be provided to the advertisement broker.
  • a recommender system can consider the sentiment map of a media content item that was enjoyed by a user. By comparing this sentiment map with sentiment maps of other content items, the recommender can choose a content item whose sentiment map is similar to the sentiment map of the content item enjoyed by the user. (When comparing, the recommender can also consult a preference profile for the user). The recommender then recommends this other content item to the user.
  • Figure 1 is an overview of an exemplary environment in which the present invention may be practiced
  • Figure 2 is a generalized schematic of some of the devices shown in Figure 1 ;
  • Figure 3 is a flowchart of an exemplary method for creating a sentiment map for a media content item
  • Figure 4 is a flowchart of an exemplary method for using a sentiment map to place advertisements.
  • Figure 5 is a flowchart of an exemplary method for using a sentiment map to recommend a media content item.
  • FIG. 1 Aspects of the present invention may be practiced in the representative communications environment 100 of Figure 1.
  • media content providers e.g., cable television head-end servers and the like
  • other servers such as media analyzers 104, advertisement brokers 106, and recommender systems 108.
  • the functions of these servers are discussed below.
  • the servers 104, 106, 108 provide, via the networking technologies 102, sentiment analysis of media content and related services to end-user devices.
  • a set-top box 1 14 generally receives television programming from various media content providers and provides a user interface (e.g., an interactive program guide) for selecting and viewing content from the cable provider.
  • a digital video recorder (not shown) can store programming for later viewing.
  • Video content may be viewed on a television monitor 1 16.
  • a laptop computer 1 18 can access both television content and web-based services either wirelessly or via the wireline network 1 12.
  • a home gateway, kiosk, digital sign, or media-restreaming device are other possible end-user devices.
  • a media-restreaming device transfers content between disparate types of networks. For example, it receives content from a cable television system 1 12 and then transmits that content over a local radio link such as WiFi to the cellular telephone 1 10.
  • the media-restreaming device usually operates in both directions to carry messages between the networks.
  • aspects of the present invention are practiced by a media-restreaming device.
  • Television programming can also be delivered to non-traditional subscriber devices such as the cellular telephone 1 10.
  • This telephone 1 10 communicates wirelessly to a wireless base station (not shown but known in the art) to access the public switched telephone network, the Internet, or other networks to access web-based services as well as the television-delivery services provided by the media content providers.
  • Wireless and wireline network technologies generally support two-way traffic: Media content and related information are delivered to the end-user devices 1 10, 1 14, 1 16, 1 18, and requests go "up" to the servers 104, 106, 108.
  • a typical user may split his attention by interacting with any or all of the end- user devices 1 10, 1 14, 1 16, 1 18 at roughly the same time or in a temporally overlapping manner. Examples in the present discussion usually assume that the user is watching the television monitor 1 16 and possibly interacting with it through the set-top box 1 14. In some situations, the user at least occasionally gives some of his attention to a "companion device" such as the cellular telephone 1 10.
  • a media analysis application analyzes the television program (possibly before the program is delivered to the user or possibly in real time) for sentiments.
  • the media analysis application produces a sentiment map of the television program.
  • the map lists segments of the program along with sentiments (e.g., "happy,” “exciting,” unknown) associated with the segments.
  • sentiments e.g., "happy,” “exciting,” unknown
  • the present discussion assumes that the media analysis application is fully embodied on one device, but in other embodiments this application can reside at least partially within the head-end of a cable provider, on a web server 104, on an end-user device such as the cellular telephone 1 10 or set- top box 1 14, or on some combination of these.
  • the sentiment map is then made available to services such as an advertisement broker 106.
  • the advertisement broker 106 determines, for example, that the next 30 seconds of the television program are "exciting.”
  • the advertisement broker 106 finds an advertisement whose sponsor wishes the advertisement to be associated with "exciting" content.
  • the advertisement broker 106 then places a bid to place that advertisement either on the television monitor 1 16 or on the user's companion device 1 10. If the bid is accepted, then the advertisement is placed temporally near the exciting segment of the television program, to the satisfaction of the advertisement sponsor. (In some situations, the advertisement is delivered to the set-top box 1 14, and the set- top box 1 14 delivers the advertisement to the television monitor 1 16.
  • connection options are well known in the art and need not be further discussed.
  • Figure 2 shows the major components of a representative server 104, 106, 108 or end-user device 1 10, 1 14, 1 18.
  • Network interfaces (also called transceivers) 200 send and receive media presentations and messages such as the sentiment map.
  • a processor 202 controls the operations of the device and, in particular, supports aspects of the present invention as illustrated in Figures 3 through 5, discussed below.
  • the user interface 204 supports a user's (or administrator's) interactions with the device. Specific uses of these components by specific devices are discussed as appropriate below.
  • Figure 3 presents a method for creating a sentiment map.
  • the method begins in step 300 when the media analyzer application receives a media content item.
  • media content item is meant very broadly: It can be a television program or movie, but it could also be a sound-only clip, a message of any kind (e.g., an e-mail with attached or embedded content), a telephone call with or without accompanying video, or even an advertisement or an interactive computer game.
  • the media content item is a television program presented on the television monitor 1 16, but all these other possibilities should be kept in mind.
  • “Receiving” encompasses many possibilities. If the media analyzer is embodied on a network server 104, then it can download the full content item and process it according to the remaining steps in Figure 3. This "offline" method has several advantages. First, the media analyzer 104 can take whatever time it needs to perform the analysis. Second, because the media analyzer 104 can review the entire content item, it can better estimate the beginning and end of a particular segment. (See the detailed description of delimitation that accompanies step 304 below.) Third, a network server 104 can perform the sentiment analysis once for a given content item and then provide the map as needed, rather than having each recipient of the content item perform its own analysis. Most of these advantages also apply if the analysis is done offline by a local user device, e.g., by the user's laptop computer 1 18 analyzing a content item stored on the user's digital video recorder.
  • the content item may be a live event streamed to a user's television monitor 1 16. Even if the content item is not actually a live broadcast, if it is being shown for the first time (e.g., the newest episode of a television series), then it is unlikely that the media analyzer will be allowed access to all of the content item before it is sent to users. In these situations, the media analyzer attempts to create the sentiment map in "real time," that is, while the content item is being received.
  • the media analyzer application runs on the user's set- top box 1 14, and it analyzes the programming as it is being received from the cable system 1 12 and then sent to the television monitor 1 16.
  • the programming can be buffered and delayed for a few seconds by the set-top box 1 14 to allow the media analysis to "keep ahead" of the point in the content item currently being viewed.
  • next two steps, 302 and 304 are, in most embodiments, inseparable.
  • the present discussion attempts, as far as possible, to present these two steps independently, but it should be kept in mind that the detailed discussion of each step informs that of the other.
  • Step 302 the media content item is analyzed for sentiments.
  • every point in time of the content item is associated with one or more sentiment-state keywords that express the mood of the content item at that point in time.
  • Step 304 is a recognition that, generally, the "point in time” is actually a relatively short temporal segment of the content item. For example, the segment from 20 seconds into the content item to 30 seconds is considered to be "happy," and the segment from 53 seconds to 1 minute, 20 seconds is "exciting.”
  • the sentiments associated with a segment and the delimitation of the segment are interrelated. In fact, it is usually the sentiment analysis itself that determines the delimitation of the segments.
  • the surrounding time in the content item can be analyzed to determine the approximate temporal extent of this happiness. That temporal extent then defines the limits of the segment that is associated with happiness.
  • metadata are available with the content item that help in delimiting the segments (e.g., a scene list with start and stop times).
  • Soundtracks generally provide distinct cues to viewers to know what sentiment is expected (e.g., a low, slow cadence in a minor key is usually associated with grief or loss by human hearers), and well known tools are available to the media analyzer to extract this information.
  • the words being said, and how they are said, also often contain clear sentiment cues.
  • the video itself may contain cues, such as the amount of time between cuts (exciting scenes usually cut very often), light levels, the amount of shadowing of a speaker's face, and how the main characters are placed with respect to one another.
  • the media analysis application can use metadata associated with the media content item such as close-captioning information, an electronic program guide listing, or a script with staging instructions.
  • a sophisticated media analysis application can mine further sources of information.
  • One interesting possibility considers "social-networking" metadata, that is, online posts and commentary produced by viewers of the media content item. These comments are often produced while the content item is being viewed. Other comments are posted later. All of these comments can be reviewed in an attempt to refine the sentiment analysis.
  • the analysis application reviews the video and audio of a segment and, based on that review, associates an "exciting" keyword with that segment.
  • a scan of online posts reveals a surprising number of viewers who found this same segment to be poorly conceived and woodenly acted. These viewers were disappointed and bored with the segment.
  • the analysis application can take these reviews into account by downgrading a confidence value (see the discussion of step 306 below) associated with the "exciting" keyword or even by assigning both an "exciting" keyword and a “boring" keyword to the same segment, the former indicating the director's intent, and the latter indicating the result actually achieved.
  • the media analysis application can attempt to map a particular user's expected response to segments of the media content item. (This is very specific as compared to the above discussion of online posts, where the posts reflect the responses of the general population to a segment.)
  • the information specific to a particular user can include, for example, a preference profile of the user, purchasing and other behavioral information, demographic information, and even current situational information such as the presence of friends and family during the viewing of the content item. All of this information can be considered when making a sentiment map personalized to this user.
  • the personalized sentiment map may tag a segment as "exciting" because this particular user is known to be an enthusiast for dog shows. The same segment may not be not tagged as exciting in a non-personalized sentiment map if the general public does not share this enthusiasm.
  • the entire media content item is made up of only one segment. If, for example, the content item is very short, such as a 30-second advertisement, then it may present only one sentiment state throughout. Generally, however, several segments can be defined within one content item. The segments may even overlap as, for example, when, partway through a "happy" segment, an "exciting" segment begins. [0041]
  • the delimitation produced by the media analysis application is often imprecise. To address this possibility, sophisticated embodiments attach to each delimitation a confidence value. For example: "It is 90% probable that this happy segment lasts for at least the next 20 seconds, and 65% probable that it continues for a further 15 seconds after that.”
  • the delimited segments do not encompass the entire content item. Some portions of the content item may simply not express a sentiment, or the media analysis application cannot discover the particular sentiment intended.
  • one or more sentiment-state keywords are associated with delimited segments. Any types of keywords can be used here including, for example, “happy,” “sad,” “exciting,” “boring,” “funny,” “romantic,” “violent,” “successful,” and the like. (As noted just above, the media analysis application may not be able to associate any sentiment-state keyword with a particular segment. In some embodiments, the keyword "unknown” is then used.) As the tools available to the media analysis application improve, it is expected that the list of keywords will increase.
  • sentiment map can include confidence values for each associated sentiment keyword and the amplitude (if any) of the keywords.
  • Step 308 stores the information generated by the evaluation and delimitation in a sentiment map. Any number of types of data representation are possible here. It should be noted that overall, the map is very much smaller than the media content item itself.
  • the map can be sent to applications that need it (step 310). (See the discussion of example applications accompanying Figure 4 and 5 below.) If the map is needed in real time, then parts of it can be sent out to waiting applications as they become available (e.g., segment by segment).
  • step 312 the sentiment mapping of a segment is revised based on further information, presumably information that was not available when the original sentiment map was created for the segment.
  • the online social-network posts, mentioned above in relation to step 302, are especially relevant here. Because some applications may need to use the sentiment map in real time ( Figure 4 gives an example), the media analyzer produces an initial sentiment map as quickly as it can with whatever information is at hand. Some online comments may be posted too late to be considered for this realtime analysis, but they can be used in the re-evaluation of step 312.
  • Portions of the media content item itself may also count as "further information" for the purposes of step 312.
  • offline processing has the advantage that it can view the entire media content item when deciding how to delimit each segment.
  • a real-time media analyzer does not have that option, but, after receiving more or even all of the content item, it can achieve much of the results of offline processing by re-considering the delimitation of segments as a result of the re-evaluation of step 312.
  • Figure 4 presents a method whereby an advertisement broker 106 uses the sentiment map when deciding where to place advertisements.
  • the method begins in step 400 when the advertisement broker 106 receives the sentiment map.
  • this may be a map of the entire media content item, produced offline by, say, a media analysis server 104.
  • the advertisement broker 106 receives the map segment by segment as produced in real time, as, for example, the content item is being distributed to viewers via the cable television system 1 12.
  • the advertisement broker 106 in step 402 compares the sentiment keywords and delimitation of a segment with a candidate advertisement.
  • the sponsor of this particular advertisement may have told the advertisement broker 106 that this advertisement should be shown in conjunction with "exciting" segments of the media content item.
  • the sponsor did not so inform the advertisement broker 106 of its intent. Instead, a sentiment map is created for the advertisement itself (as noted above in relation to step 304, this map may only contain a single segment). That is, it is recognized that an advertisement is also a media content item in its own right and can be subjected to the same type of analysis described above in reference to Figure 3. In this case, the analysis may show that the advertisement is both "exciting" and "upbeat.” The advertisement broker 106 can then infer that this advertisement would be best received if it were shown during "exciting" or "upbeat” segments of the media content item that the user is watching on his television monitor 1 16.
  • the advertisement broker 106 determines which sentiments are most favorable to this advertisement. The comparison may also consider the amplitude of the sentiment and the delimitation of the segment (i.e., the segment may be too short to use with this advertisement). Confidence values in the map can also be considered: If the content item is drawing to a close, and if the advertisement broker 106 is tasked with presenting a given number of advertisements during the presentation of the content item, then the advertisement broker 106 may have to settle for a less than ideal, or for a less confidently ideal, segment for the advertisement. If a match is favorable enough (usually as defined by the advertisement sponsor or as inferred by the advertisement broker 106), then the advertisement broker 106 proceeds to step 404a where it attempts to place this advertisement in conjunction with this segment
  • the advertisement broker 106 submits a bid to place the advertisement at a given time.
  • the bid can include a proposed fee to be paid to the content provider, and can specify whether the advertisement is to be placed within the stream of the media content item itself (as viewed on the television monitor 1 16 to continue the pervasive example of this discussion) or on a companion device, such as the user's cellular telephone 1 10.
  • a companion device such as the user's cellular telephone 1 10.
  • well known methods can be applied to determine whether or not the user has a companion device, whether or not that device is turned on, and whether or not the user is interacting with that device (indicating that the companion device has captured at least some of the user's attention). If the bid is accepted, then the advertisement is placed accordingly.
  • step 312 of Figure 3 the sentiment map is redrawn as more information becomes available to the media analysis application.
  • step 406 of Figure 4 the advertisement broker 106 reviews the revised map. If a bid was placed and accepted, but the revised map shows that the segment was not all it was thought to be (e.g., based on viewer's online responses, a supposedly "romantic" scene fell flat), then the advertisement broker 106 can request a refund of part of the bid price. The possibility of a refund could make advertisement brokers 106 more willing to place a reasonable amount of reliance on sentiment mapping and on placing bids based on these maps.
  • Figure 5 presents a recommender system 108 that uses sentiment maps.
  • the recommender 108 receives sentiment maps of two media content items. The maps are compared in step 504, and, if the comparison is favorable in some sense, then the two content items are associated with one another in step 506.
  • This is a generalization of the special case discussed above in relation to step 402 of Figure 4 where a sentiment map of an advertisement is compared against the map of a content item to see where the advertisement should be placed.
  • sentiment maps can be used to categorize content items in a more meaningful fashion than categorization by actors, setting, or plot elements. If a user is known to like one content item, then a second content item with a similar sentiment map can be recommended to him in step 508. The user's reaction to the recommendation may help to improve the quality of the sentiment mapping.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

La présente invention se rapporte à un procédé adapté pour évaluer les « états de sentiment » d'un élément de contenu multimédia (302). En d'autres termes, des segments de l'élément de contenu sont déterminés comme désignant, par exemple, un sentiment de « joie », d'« excitation », de « tristesse », d'« amusement », et similaires. Une « carte de sentiments » est créée (308). Cette carte délimite (304) des segments de l'élément de contenu et contient les mots-clés qui désignent les états de sentiment associés (306) aux segments. Certaines cartes comprennent une amplitude pour chaque mot-clé attribué à un état de sentiment, et une valeur de confiance attribuée à la délimitation des segments et à chaque mot-clé. Dans un mode de réalisation de la carte de sentiments, qui est fourni à titre d'exemple de l'invention, un courtier en publicité met en correspondance (404a) les sentiments de ses offres publicitaires par rapport à des segments d'un élément de contenu, dans le but de placer des publicités adéquates à des moments où leur perception sera la plus favorable. Dans un autre exemple, un système de recommandation recommande (508) à un utilisateur un élément de contenu dont la carte de sentiments coïncide (506) favorablement avec celle d'un élément de contenu déjà apprécié par l'utilisateur.
EP13708972.8A 2012-03-14 2013-02-21 Mise en correspondance de sentiments dans un élément de contenu multimédia Ceased EP2826254A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US13/419,615 US9106979B2 (en) 2012-03-14 2012-03-14 Sentiment mapping in a media content item
PCT/US2013/027049 WO2013138038A1 (fr) 2012-03-14 2013-02-21 Mise en correspondance de sentiments dans un élément de contenu multimédia

Publications (1)

Publication Number Publication Date
EP2826254A1 true EP2826254A1 (fr) 2015-01-21

Family

ID=47846159

Family Applications (1)

Application Number Title Priority Date Filing Date
EP13708972.8A Ceased EP2826254A1 (fr) 2012-03-14 2013-02-21 Mise en correspondance de sentiments dans un élément de contenu multimédia

Country Status (5)

Country Link
US (2) US9106979B2 (fr)
EP (1) EP2826254A1 (fr)
KR (1) KR101696988B1 (fr)
CA (1) CA2867019C (fr)
WO (1) WO2013138038A1 (fr)

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8995822B2 (en) 2012-03-14 2015-03-31 General Instrument Corporation Sentiment mapping in a media content item
US10681427B2 (en) 2012-03-14 2020-06-09 Arris Enterprises Llc Sentiment mapping in a media content item
CN104428804A (zh) * 2012-05-04 2015-03-18 B-Sm@Rk公司 用于评价对象的方法和装置
US10097882B2 (en) 2015-08-11 2018-10-09 Arris Enterprises Llc Back-end content analysis system to initiate second-screen confirmation
US9628839B1 (en) 2015-10-06 2017-04-18 Arris Enterprises, Inc. Gateway multi-view video stream processing for second-screen content overlay
CN105825044A (zh) * 2016-03-10 2016-08-03 深圳市前海安测信息技术有限公司 基于网页浏览的医疗大数据分析及处理系统及方法
WO2017209482A1 (fr) * 2016-05-31 2017-12-07 (주)잼투고 Système et procédé de fourniture d'un contenu vidéo
US10043062B2 (en) 2016-07-13 2018-08-07 International Business Machines Corporation Generating auxiliary information for a media presentation
US9741258B1 (en) 2016-07-13 2017-08-22 International Business Machines Corporation Conditional provisioning of auxiliary information with a media presentation
US11397740B2 (en) * 2017-07-24 2022-07-26 Mycelebs Co., Ltd. Method and apparatus for providing information by using degree of association between reserved word and attribute language
US10636449B2 (en) * 2017-11-06 2020-04-28 International Business Machines Corporation Dynamic generation of videos based on emotion and sentiment recognition
JP6368025B2 (ja) * 2017-12-14 2018-08-01 ヤフー株式会社 装置、方法及びプログラム
JP6568263B2 (ja) * 2018-04-27 2019-08-28 ヤフー株式会社 装置、方法及びプログラム
US11250468B2 (en) 2019-02-28 2022-02-15 International Business Machines Corporation Prompting web-based user interaction
US11645578B2 (en) 2019-11-18 2023-05-09 International Business Machines Corporation Interactive content mobility and open world movie production
US20220020061A1 (en) * 2020-07-14 2022-01-20 At&T Intellectual Property I, L.P. Apparatuses and methods for populating inventory associated with content items in accordance with emotionally guided placements and adaptations
US11797753B2 (en) * 2022-06-27 2023-10-24 Univerzita Palackého v Olomouci System and method for adapting text-based data structures to text samples

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020099518A1 (en) * 2001-01-23 2002-07-25 Tovinkere Vasanth R. Method and system for detecting semantic events
US20060200342A1 (en) * 2005-03-01 2006-09-07 Microsoft Corporation System for processing sentiment-bearing text
US20080262912A1 (en) * 2007-04-20 2008-10-23 Ullas Gargi Media Advertising
WO2010078650A1 (fr) * 2009-01-07 2010-07-15 Jigsee Inc. Identification, recommandation et fourniture d'un contenu média approprié
WO2011009101A1 (fr) * 2009-07-16 2011-01-20 Bluefin Lab, Inc. Estimation et affichage de l’intérêt collectif pour des médias temporels

Family Cites Families (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6002831A (en) 1995-05-16 1999-12-14 Hitachi, Ltd. Image recording/reproducing apparatus
JP2005018734A (ja) * 2003-04-30 2005-01-20 Seiko Epson Corp ファイル管理方法及びその装置、並びにアノテーション情報生成方法及びその装置
US7979877B2 (en) 2003-12-23 2011-07-12 Intellocity Usa Inc. Advertising methods for advertising time slots and embedded objects
US20060212897A1 (en) 2005-03-18 2006-09-21 Microsoft Corporation System and method for utilizing the content of audio/video files to select advertising content for display
WO2007043679A1 (fr) * 2005-10-14 2007-04-19 Sharp Kabushiki Kaisha Dispositif et programme de traitement d’informations
US7788131B2 (en) 2005-12-15 2010-08-31 Microsoft Corporation Advertising keyword cross-selling
US7546532B1 (en) 2006-02-17 2009-06-09 Adobe Systems Incorporated Methods and apparatus for editing content
US20090070683A1 (en) 2006-05-05 2009-03-12 Miles Ward Consumer-generated media influence and sentiment determination
KR100828371B1 (ko) 2006-10-27 2008-05-08 삼성전자주식회사 컨텐츠의 메타 데이터 생성 방법 및 장치
US8090613B2 (en) 2007-12-10 2012-01-03 Kalb Kenneth J System and method for real-time management and optimization of off-line advertising campaigns
US7889073B2 (en) 2008-01-31 2011-02-15 Sony Computer Entertainment America Llc Laugh detector and system and method for tracking an emotional response to a media presentation
US20090217319A1 (en) 2008-02-22 2009-08-27 Weiss Jonathan B Method and system for providing targeted television advertising
US20100042421A1 (en) 2008-08-18 2010-02-18 Microsoft Corporation Context based advertisement bidding mechanism
US20100169326A1 (en) * 2008-12-31 2010-07-01 Nokia Corporation Method, apparatus and computer program product for providing analysis and visualization of content items association
KR20100095924A (ko) 2009-02-23 2010-09-01 삼성전자주식회사 동영상의 상황정보를 반영한 광고 키워드 추출 방법 및 장치
US8166032B2 (en) * 2009-04-09 2012-04-24 MarketChorus, Inc. System and method for sentiment-based text classification and relevancy ranking
JP5113796B2 (ja) * 2009-04-15 2013-01-09 ヤフー株式会社 感情マッチング装置、感情マッチング方法、及びプログラム
US9400982B2 (en) 2009-09-29 2016-07-26 Verizon Patent And Licensing Inc. Real time television advertisement shaping
CN102486771B (zh) 2009-11-30 2015-07-08 国际商业机器公司 在网页上发布指定内容的方法和系统
US9201863B2 (en) 2009-12-24 2015-12-01 Woodwire, Inc. Sentiment analysis from social media content
US20110197220A1 (en) 2010-02-09 2011-08-11 Google Inc. Customized television advertising
US20120143693A1 (en) 2010-12-02 2012-06-07 Microsoft Corporation Targeting Advertisements Based on Emotion
US8995822B2 (en) 2012-03-14 2015-03-31 General Instrument Corporation Sentiment mapping in a media content item
US10681427B2 (en) 2012-03-14 2020-06-09 Arris Enterprises Llc Sentiment mapping in a media content item

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020099518A1 (en) * 2001-01-23 2002-07-25 Tovinkere Vasanth R. Method and system for detecting semantic events
US20060200342A1 (en) * 2005-03-01 2006-09-07 Microsoft Corporation System for processing sentiment-bearing text
US20080262912A1 (en) * 2007-04-20 2008-10-23 Ullas Gargi Media Advertising
WO2010078650A1 (fr) * 2009-01-07 2010-07-15 Jigsee Inc. Identification, recommandation et fourniture d'un contenu média approprié
WO2011009101A1 (fr) * 2009-07-16 2011-01-20 Bluefin Lab, Inc. Estimation et affichage de l’intérêt collectif pour des médias temporels

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
A. HANJALIC: "Extracting moods from pictures and sounds: towards truly personalized TV", IEEE SIGNAL PROCESSING MAGAZINE, vol. 23, no. 2, 1 March 2006 (2006-03-01), pages 90 - 100, XP055059619, ISSN: 1053-5888, DOI: 10.1109/MSP.2006.1621452 *
See also references of WO2013138038A1 *
SHILIANG ZHANG ET AL: "Utilizing affective analysis for efficient movie browsing", IMAGE PROCESSING (ICIP), 2009 16TH IEEE INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 7 November 2009 (2009-11-07), pages 1853 - 1856, XP031628307, ISBN: 978-1-4244-5653-6 *

Also Published As

Publication number Publication date
CA2867019A1 (fr) 2013-09-19
KR101696988B1 (ko) 2017-01-16
US20150331954A1 (en) 2015-11-19
WO2013138038A1 (fr) 2013-09-19
CA2867019C (fr) 2018-05-01
US9106979B2 (en) 2015-08-11
KR20140139549A (ko) 2014-12-05
US20130246447A1 (en) 2013-09-19

Similar Documents

Publication Publication Date Title
CA2867019C (fr) Mise en correspondance de sentiments dans un element de contenu multimedia
US11252481B2 (en) Sentiment mapping in a media content item
US8995822B2 (en) Sentiment mapping in a media content item
US11587128B2 (en) Verifying presentation of video content
US11134299B2 (en) Selection and presentation of context-relevant supplemental content and advertising
US10032187B2 (en) Device functionality-based content selection
US10299015B1 (en) Time-based content presentation
US20170323307A1 (en) Specifying media content placement criteria
RU2539585C2 (ru) Адаптивное размещение вспомогательных медиаданных в рекомендательных системах
US20150016802A1 (en) Goal-based video delivery system
WO2007104026A2 (fr) Sélection et présentation de contenus et d'annonces publicitaires complémentaires liés au contexte
AU2011203560B2 (en) Improved advertising with video ad creatives

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20141014

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

DAX Request for extension of the european patent (deleted)
RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: ARRIS TECHNOLOGY, INC.

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: ARRIS ENTERPRISES, INC.

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20161209

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: ARRIS ENTERPRISES LLC

APBK Appeal reference recorded

Free format text: ORIGINAL CODE: EPIDOSNREFNE

APBN Date of receipt of notice of appeal recorded

Free format text: ORIGINAL CODE: EPIDOSNNOA2E

APBR Date of receipt of statement of grounds of appeal recorded

Free format text: ORIGINAL CODE: EPIDOSNNOA3E

APAF Appeal reference modified

Free format text: ORIGINAL CODE: EPIDOSCREFNE

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: ARRIS INTERNATIONAL IP LTD

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: BISON PATENT LICENSING, LLC

REG Reference to a national code

Ref country code: DE

Ref legal event code: R003

APBT Appeal procedure closed

Free format text: ORIGINAL CODE: EPIDOSNNOA9E

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED

18R Application refused

Effective date: 20231006